CN109933431A - A kind of intelligent client load equalization methods and system - Google Patents
A kind of intelligent client load equalization methods and system Download PDFInfo
- Publication number
- CN109933431A CN109933431A CN201910180733.0A CN201910180733A CN109933431A CN 109933431 A CN109933431 A CN 109933431A CN 201910180733 A CN201910180733 A CN 201910180733A CN 109933431 A CN109933431 A CN 109933431A
- Authority
- CN
- China
- Prior art keywords
- service
- application server
- isp
- weight
- intelligent client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Computer And Data Communications (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of intelligent client load equalization methods and system, belong to computer field, the technical problem to be solved in the present invention is how to be comprehensively considered according to the characteristics of each service, accomplish appropriate selection load balancing, a kind of technical solution are as follows: 1. intelligent client load equalization methods, steps are as follows: S1, service response performance evaluation: recording service response time when each service call, the score of ISP is calculated according to service response time;S2, synchronous service load balancing: performance distributes weight according to performance scores as load foundation to synchronous service according to response;S3, asynchronous service load balancing.2., a kind of equal balance system of intelligent client load, including service number of concurrent statistical analysis center, service caller application server and ISP's application server, service caller application server and ISP's application server be arranged in a one-to-one correspondence.
Description
Technical field
The present invention relates to field of computer technology, specifically a kind of intelligent client load equalization methods and it is
System.
Background technique
Under micro services framework, in order to realize the maximization of performance, application server extending transversely is common deployment side
Formula, the same service have multiple suppliers.When client call micro services, just there are application server select permeability, client
Service is balanced to provide this ability, can be according to one application server of policy selection of configuration, and there are many loads for industry
Balance policy, for example, at random, poll, weight, response time etc., every kind of strategy has limitation, such as due to each service mentions
The hardware of donor may be not quite similar, and random and poll can not identify this difference, although weight can identify hardware performance
Difference, but performance consumed by each service also has very big difference, or even is not in an order of magnitude, the response time may be right
Part synchronous service is effective, and asynchronous service does not use then, therefore how to be comprehensively considered according to the characteristics of each service, accomplishes
Appropriate selection load balancing is technical problem urgently to be solved in currently available technology.
The patent document of Patent No. CN103957251A discloses a kind of method for realizing server load balancing, the party
Method specifically includes that the initial ability value of setting server, and is asked according to the authentication business access that initial ability value distributes client
It asks;According to the interaction time length that client certificate business access is requested, the self-learning capability value of server is set, action of going forward side by side
State adjustment;When client initiates authentication business access request, according to the initial ability value of each server and
The authentication business access request that client is initiated is assigned to corresponding server by the addition summation of self-learning capability value.The skill
Art scheme introduces the concept of server-capabilities value intelligently to distribute business flowing of access, solves the fixed active-standby mode of tradition
The problem of lower separate unit server load is overweight, cannot reasonably distribute business data traffic, enables server to be in for a long time
Steady efficient working condition guarantees that customer service runs well;But the technical solution cannot be according to each service the characteristics of
Comprehensively considered, accomplishes appropriate selection load balancing.
The intelligent dynamic that the patent document of Patent No. CN102404390B discloses a kind of high-speed real-time library is negative
Equalization methods are carried, this method includes the following steps: to configure the consistent collection of content in each client and cluster server node
Group's arbitral table;Back-end data memory module is divided into different physical store subregions;By client and each cluster server
Node establishes communication connection;Server end cluster quorum on each each cluster server node of cluster server node maintenance
Table content is consistent;Receive and process the task that client sends over;Client is initiated data to global virtual ip address and is read
The request write;Network communication management module finally sends toward mesh cluster server node;Target cluster server node is connecing
After receiving the request that client submission comes, necessary verification is carried out, then legal request is handled.But the technical side
Case cannot be comprehensively considered according to the characteristics of each service, accomplish appropriate selection load balancing.
Summary of the invention
Technical assignment of the invention is to provide a kind of intelligent client load equalization methods and system, come solve how root
The problem of being comprehensively considered according to the characteristics of each service, accomplishing appropriate selection load balancing.
Technical assignment of the invention realizes in the following manner, a kind of intelligent client load equalization methods, step
It is as follows:
S1, service response performance evaluation: recording service response time when each service call, according to service response time meter
Calculate the score of ISP;
S2, synchronous service load balancing: performance is as load foundation according to response for synchronous service, according to performance scores
Distribute weight;
S3, asynchronous service load balancing: weight is obtained according to the ability value of application server and corresponding number of concurrent.
Preferably, the information of the service response time includes ISP ID, service ID, service is time-consuming and takes
Business allocating time.
More preferably, the specific method is as follows for the service response performance evaluation:
S101, the service cluster average response time T for calculating ISP ID1With the average response time of service ID
T2;
S102, calculate ISP score T calculation formula:
T=T1-T2;
Wherein, the score T value of ISP is bigger, illustrates that the performance of ISP is better.
More preferably, performance scores are higher in the step S2, and weight is bigger, otherwise weight is smaller;Weight is bigger, is selected
Probability it is higher.
More preferably, specific step is as follows for the step S3 asynchronous service load balancing:
After S301, each service request, record concurrently counts to memory;
S302, statistics center is reported to every the specified time, end is called periodically to draw from statistics center every specified time
Take number of concurrent;
S303, obtain weight according to the ability value and corresponding number of concurrent of application server, ability is bigger and number of concurrent more
Small, weight is higher, and the probability being chosen to is higher.
More preferably, the Distribution Indexes weight of synchronous service time according to response, asynchronous service is according to concurrency index
Distribute weight.
A kind of equal balance system of intelligent client load, the system include service number of concurrent statistical analysis center, service tune
User's application server and ISP's application server, service caller application server and ISP's application service
Device is arranged in a one-to-one correspondence, and service number of concurrent statistical analysis center timing pulls the information of service caller server, is serviced simultaneously
Caller application server timing reports information to service number of concurrent statistical analysis center;Service caller server transmission path by
Strategy arrives ISP's application server.
Preferably, the implementation process of the system is specific as follows:
(1), service type is arranged: setting service is synchronous service or asynchronous service;
(2), each application server hardware configuration is set or hardware configuration score is directly set;
(3), setting service consumption holds equal response events measurement period level with both hands;
(4), number of concurrent statistical report is set, pulls the period.
More preferably, service type is defaulted as asynchronous service in the step (1).
More preferably, in the step (2) application server hardware configuration include memory (G), CPU number and hard disk whether
For solid state hard disk.
The client load equalization methods and system of intelligence of the invention have the advantage that
(1) present invention is under micro services framework, and when client call micro services, load balancing is fixed, Bu Nenggen
Suitable load strategy is dynamically selected according to actual conditions, and is no longer fixed using load strategy of the present invention, it can be according to reality
When service feature and performance data, intelligently select suitable load strategy, and then improve the handling capacity and performance of system, have
There is application value well;
(2), synchronous service load balancing of the invention be using synchronous service according to response performance as load according to
According to according to performance scores distribution weight, performance scores are higher, and weight is bigger, otherwise weight is smaller;Weight is bigger, selected
Probability is higher;
(3), the concurrency index of asynchronous service is concurrency/CPU number, and concurrency is higher, and assigned probability is got over
Small, formula can also be customized;
(4), service load balancing is that client load is balanced, the Distribution Indexes weight of synchronous service time according to response,
Asynchronous service is according to concurrency Distribution Indexes weight.
Detailed description of the invention
The following further describes the present invention with reference to the drawings.
Attached drawing 1 is the structural block diagram of the equal balance system of client load of intelligence.
Specific embodiment
To a kind of intelligent client load equalization methods of the invention and it is referring to Figure of description and specific embodiment
System is described in detail below.
Embodiment 1:
The client load equalization methods of intelligence of the invention, steps are as follows:
S1, service response performance evaluation: recording service response time when each service call, according to service response time meter
Calculate the score of ISP;Wherein, the information of service response time includes ISP ID, service ID, service time-consuming
And the service call time.
The specific method is as follows for service response performance evaluation:
S101, the service cluster average response time T for calculating ISP ID1With the average response time of service ID
T2;
S102, calculate ISP score T calculation formula:
T=T1-T2;
Wherein, the score T value of ISP is bigger, illustrates that the performance of ISP is better.
S2, synchronous service load balancing: performance is as load foundation according to response for synchronous service, according to performance scores
Distribute weight;Performance scores are higher, and weight is bigger, otherwise weight is smaller;Weight is bigger, and selected probability is higher.
S3, asynchronous service load balancing: weight is obtained according to the ability value of application server and corresponding number of concurrent;
Wherein, specific step is as follows for asynchronous service load balancing:
After S301, each service request, record concurrently counts to memory;
S302, statistics center is reported to every the specified time, end is called periodically to draw from statistics center every specified time
Take number of concurrent;
S303, obtain weight according to the ability value and corresponding number of concurrent of application server, ability is bigger and number of concurrent more
Small, weight is higher, and the probability being chosen to is higher.
Wherein, asynchronous service due on the response time difference it is little, cannot again using the response time as load foundation, this
Invention is CPU as foundation, such as application server hardware configuration according to the ability value of application server and the size of number of concurrent
For 2 cores, it is interior save as 8G, ability value is obtained according to calculation formula;After each service request, records concurrently count to memory first, so
Statistics center is reported to every the specified time afterwards, end is called periodically to pull number of concurrent from statistics center every specified time, then
Weight is obtained according to the ability value of application server and corresponding number of concurrent, ability is bigger, number of concurrent is smaller, and weight is higher, quilt
The probability chosen is higher.
Wherein, the Distribution Indexes weight of synchronous service time according to response, asynchronous service are weighed according to concurrency Distribution Indexes
Value.
Embodiment 2:
The equal balance system of client load of intelligence of the invention, the system include service number of concurrent statistical analysis center, clothes
Business caller application server and ISP's application server, service caller application server and ISP's application
Server is arranged in a one-to-one correspondence, and service number of concurrent statistical analysis center timing pulls the information of service caller server, simultaneously
Service caller application server timing reports information to service number of concurrent statistical analysis center;The transmission of service caller server
Routing policy is to ISP's application server.
Implementation process is specific as follows:
(1), service type is arranged: setting service is synchronous service or asynchronous service;Service type is defaulted as asynchronous clothes
Business.
(2), each application server hardware configuration is set or hardware configuration score is directly set;Application server hardware is matched
It sets including whether memory (G), CPU number and hard disk are solid state hard disk.
(3), setting service consumption holds equal response events measurement period level with both hands;
(4), number of concurrent statistical report is set, pulls the period.
Embodiment 3:
As shown in Fig. 1, the equal balance system of client load of intelligence of the invention, the system include service number of concurrent statistics
Analysis center, service caller application server one, service caller application server two, ISP's application server one
With ISP's application server two, services the center timing of number of concurrent statistical analysis and pull service caller application server one
With the information of service caller application server two, while service caller application server one and service caller application clothes
Business two timing of device reports information to service number of concurrent statistical analysis center;Service caller application server one and service caller
Application server two transmits routing policy to ISP's application server one and ISP's application server two.
Implementation process is specific as follows:
(1), service type is arranged: setting service is synchronous service or asynchronous service;Service type is defaulted as asynchronous clothes
Business.
(2), each application server hardware configuration is set or hardware configuration score is directly set;Application server hardware is matched
It sets including whether memory (G), CPU number and hard disk are solid state hard disk.
(3), setting service consumption holds equal response events measurement period level with both hands, is defaulted as 2 seconds;
(4), number of concurrent statistical report is set, pulls the period, is defaulted as 2 seconds.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (10)
1. a kind of intelligent client load equalization methods, which is characterized in that steps are as follows:
S1, service response performance evaluation: service response time is recorded when each service call, is calculated according to service response time
The score of ISP;
S2, synchronous service load balancing: synchronous service according to response distribute as load foundation according to performance scores by performance
Weight;
S3, asynchronous service load balancing: weight is obtained according to the ability value of application server and corresponding number of concurrent.
2. intelligent client load equalization methods according to claim 1, which is characterized in that the service response time
Information include that ISP ID, service ID, service be time-consuming and the service call time.
3. intelligent client load equalization methods according to claim 1 or 2, which is characterized in that the service response
The specific method is as follows for performance evaluation:
S101, the service cluster average response time T for calculating ISP ID1With the average response time T of service ID2;
S102, calculate ISP score T calculation formula:
T=T1-T2;
Wherein, the score T value of ISP is bigger, illustrates that the performance of ISP is better.
4. intelligent client load equalization methods according to claim 3, which is characterized in that performance in the step S2
Score is higher, and weight is bigger, otherwise weight is smaller;Weight is bigger, and selected probability is higher.
5. intelligent client load equalization methods according to claim 4, which is characterized in that the asynchronous clothes of step S3
Being engaged in, specific step is as follows for load balancing:
After S301, each service request, record concurrently counts to memory;
S302, statistics center is reported to every the specified time, end is called periodically to pull simultaneously from statistics center every specified time
Send out number;
S303, weight is obtained according to the ability value and corresponding number of concurrent of application server, ability is bigger and number of concurrent is smaller, power
Again higher, the probability being chosen to is higher.
6. intelligent client load equalization methods according to claim 5, which is characterized in that the synchronous service according to
The Distribution Indexes weight of response time, asynchronous service is according to concurrency Distribution Indexes weight.
7. a kind of equal balance system of intelligent client load, which is characterized in that the system includes in service number of concurrent statistical analysis
The heart, service caller application server and ISP's application server, service caller application server and service provide
Person's application server is arranged in a one-to-one correspondence, and service number of concurrent statistical analysis center timing pulls the letter of service caller server
Breath, while service caller application server timing reports information to service number of concurrent statistical analysis center;Service caller clothes
Business device transmits routing policy to ISP's application server.
8. the equal balance system of intelligent client load according to claim 7, which is characterized in that the implementation process of the system
It is specific as follows:
(1), service type is arranged: setting service is synchronous service or asynchronous service;
(2), each application server hardware configuration is set or hardware configuration score is directly set;
(3), setting service consumption holds equal response events measurement period level with both hands;
(4), number of concurrent statistical report is set, pulls the period.
9. the equal balance system of intelligent client load according to claim 8, which is characterized in that clothes in the step (1)
Service type is defaulted as asynchronous service.
10. the equal balance system of client load of intelligence according to claim 8 or claim 9, which is characterized in that the step (2)
Middle application server hardware configuration includes whether memory, CPU number and hard disk are solid state hard disk.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910180733.0A CN109933431B (en) | 2019-03-11 | 2019-03-11 | Intelligent client load balancing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910180733.0A CN109933431B (en) | 2019-03-11 | 2019-03-11 | Intelligent client load balancing method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109933431A true CN109933431A (en) | 2019-06-25 |
CN109933431B CN109933431B (en) | 2023-04-04 |
Family
ID=66986742
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910180733.0A Active CN109933431B (en) | 2019-03-11 | 2019-03-11 | Intelligent client load balancing method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109933431B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110389841A (en) * | 2019-07-25 | 2019-10-29 | 中南民族大学 | A kind of server load balancing method, apparatus and storage medium |
CN110691118A (en) * | 2019-08-30 | 2020-01-14 | 许昌许继软件技术有限公司 | Service selection method and device in micro-service cluster |
CN110839086A (en) * | 2019-12-23 | 2020-02-25 | 吉林省民航机场集团公司 | High-concurrency load balancing processing method |
CN112532743A (en) * | 2020-12-18 | 2021-03-19 | 上海安畅网络科技股份有限公司 | Intelligent load balancing method and device and storage medium |
CN114048046A (en) * | 2021-11-08 | 2022-02-15 | 马上消费金融股份有限公司 | Service calling method and device and load balancing equipment |
CN114567637A (en) * | 2022-03-01 | 2022-05-31 | 浪潮云信息技术股份公司 | Method and system for intelligently setting weight of load balancing back-end server |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050027862A1 (en) * | 2003-07-18 | 2005-02-03 | Nguyen Tien Le | System and methods of cooperatively load-balancing clustered servers |
CN103049245A (en) * | 2012-10-25 | 2013-04-17 | 浪潮电子信息产业股份有限公司 | Software performance optimization method based on central processing unit (CPU) multi-core platform |
CN103945005A (en) * | 2014-05-06 | 2014-07-23 | 江苏物联网研究发展中心 | Multiple evaluation indexes based dynamic load balancing framework |
CN104579996A (en) * | 2013-10-17 | 2015-04-29 | 中国电信股份有限公司 | Cluster load balancing method and system |
CN108121312A (en) * | 2017-11-29 | 2018-06-05 | 南瑞集团有限公司 | ARV SiteServer LBSs and method based on integrated water electricity control platform |
CN108494868A (en) * | 2018-03-30 | 2018-09-04 | 三盟科技股份有限公司 | A kind of load-balancing method under the operation system based on cloud and system |
-
2019
- 2019-03-11 CN CN201910180733.0A patent/CN109933431B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050027862A1 (en) * | 2003-07-18 | 2005-02-03 | Nguyen Tien Le | System and methods of cooperatively load-balancing clustered servers |
CN103049245A (en) * | 2012-10-25 | 2013-04-17 | 浪潮电子信息产业股份有限公司 | Software performance optimization method based on central processing unit (CPU) multi-core platform |
CN104579996A (en) * | 2013-10-17 | 2015-04-29 | 中国电信股份有限公司 | Cluster load balancing method and system |
CN103945005A (en) * | 2014-05-06 | 2014-07-23 | 江苏物联网研究发展中心 | Multiple evaluation indexes based dynamic load balancing framework |
CN108121312A (en) * | 2017-11-29 | 2018-06-05 | 南瑞集团有限公司 | ARV SiteServer LBSs and method based on integrated water electricity control platform |
CN108494868A (en) * | 2018-03-30 | 2018-09-04 | 三盟科技股份有限公司 | A kind of load-balancing method under the operation system based on cloud and system |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110389841A (en) * | 2019-07-25 | 2019-10-29 | 中南民族大学 | A kind of server load balancing method, apparatus and storage medium |
CN110691118A (en) * | 2019-08-30 | 2020-01-14 | 许昌许继软件技术有限公司 | Service selection method and device in micro-service cluster |
CN110839086A (en) * | 2019-12-23 | 2020-02-25 | 吉林省民航机场集团公司 | High-concurrency load balancing processing method |
CN112532743A (en) * | 2020-12-18 | 2021-03-19 | 上海安畅网络科技股份有限公司 | Intelligent load balancing method and device and storage medium |
CN112532743B (en) * | 2020-12-18 | 2021-11-30 | 上海安畅网络科技股份有限公司 | Intelligent load balancing method and device and storage medium |
CN114048046A (en) * | 2021-11-08 | 2022-02-15 | 马上消费金融股份有限公司 | Service calling method and device and load balancing equipment |
CN114048046B (en) * | 2021-11-08 | 2022-10-11 | 马上消费金融股份有限公司 | Service calling method and device and load balancing equipment |
CN114567637A (en) * | 2022-03-01 | 2022-05-31 | 浪潮云信息技术股份公司 | Method and system for intelligently setting weight of load balancing back-end server |
Also Published As
Publication number | Publication date |
---|---|
CN109933431B (en) | 2023-04-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109933431A (en) | A kind of intelligent client load equalization methods and system | |
CN110266716B (en) | Unified service platform system of power grid | |
CN109618002B (en) | Micro-service gateway optimization method, device and storage medium | |
CN108712464A (en) | A kind of implementation method towards cluster micro services High Availabitity | |
CN103067293B (en) | Method and system for multiplex and connection management of a load balancer | |
CN109218355A (en) | Load equalizing engine, client, distributed computing system and load-balancing method | |
US9015227B2 (en) | Distributed data processing system | |
US9628556B2 (en) | Decentralized request routing | |
US11336718B2 (en) | Usage-based server load balancing | |
CN103747274B (en) | A kind of video data center setting up cache cluster and cache resources dispatching method thereof | |
CN104796422A (en) | Online customer service staff equilibrium assignment method and online customer service staff equilibrium assignment device | |
CN103699445A (en) | Task scheduling method, device and system | |
CN102611735A (en) | Load balancing method and system of application services | |
CN109271243B (en) | Cluster task management system | |
CN103401947A (en) | Method and device for allocating tasks to multiple servers | |
US10346367B1 (en) | Load shedding techniques for distributed services with persistent client connections to ensure quality of service | |
CN109412966B (en) | Large-scale log transmission method, device and system | |
CN102281190A (en) | Networking method for load balancing apparatus, server and client access method | |
CN106572181A (en) | Object storage interface load balancing method and system based on cluster file system | |
CN113312160A (en) | Techniques for behavioral pairing in a task distribution system | |
CN108259421A (en) | The statistical method and system of a kind of user activity | |
CN109639796A (en) | A kind of implementation of load balancing, device, equipment and readable storage medium storing program for executing | |
CN111258760A (en) | Platform management method, system, device and storage medium | |
CN110515728B (en) | Server scheduling method and device, electronic equipment and machine-readable storage medium | |
CN108111567A (en) | Realize the uniform method and system of server load |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |